High prevalence of late presentation with advanced HIV disease and its predictors among newly diagnosed patients in Kumasi, Ghana

Background Late presentation with advanced HIV disease (LP-AHD) remains a significant challenge to Human Immunodeficiency Virus (HIV) care, contributing to increased morbidity, mortality, and healthcare costs. Despite global efforts to enhance early diagnosis, a considerable proportion of individuals with HIV infection are unaware of being infected and therefore present late for HIV care. For the first time in Ghana, this study assessed the prevalence of LP-AHD and associated factors among people diagnosed with HIV (PDWH). Method This bi-center retrospective cross-sectional study included 315 PDWH at the Aniniwah Medical Centre and Komfo Anokye Teaching Hospital, both in Kumasi, Ghana. A well-structured questionnaire was used to collect data on sociodemographic, clinical, lifestyle and psychosocial factors from the study participants. Statistical analyses were done in SPSS version 26.0 and GraphPad Prism version 8.0 at significant p-value of < 0.05 and 95% confidence interval. Predictors of LP-AHD were assessed using binary logistic regression models. Results This study observed that, 90 out of the 315 study PDWH (28.6%) reported late with advanced HIV disease (AHD). Participants within the age group of 36–45 years (adjusted Odds Ratio [aOR]: 0.32, 95% CI: 0.14–0.69; p = 0.004) showed a significantly decreased likelihood of LP-AHD. However, participants who perceived cost of HIV care to be high (aOR: 7.04, 95% CI: 1.31–37.91; p = 0.023), who were diagnosed based on clinical suspicion (aOR: 13.86, 95 CI: 1.83–104.80; p = 0.011), and missed opportunities for early diagnosis by clinicians (aOR: 2.47, 95% CI: 1.30–4.74; p = 0.006) were significantly associated with increased likelihood of LP-AHD. Conclusion The prevalence of LP-AHD among PDWH in Ghana is high. Efforts to improve early initiation of HIV/AIDS care should focus on factors such as the high perceived costs of HIV care, diagnosis based on clinical suspicion, and missed opportunities for early diagnosis by physicians.


Introduction
Acquired Immune Deficiency Syndrome (AIDS) caused by Human Immunodeficiency Virus (HIV), remains a significant global health challenge despite efforts by international and local initiatives to address the epidemic [1].At the end of 2022, an estimated 39 million people worldwide were living with HIV [2].Africa remains the most affected, with nearly 1 in every 25 adults (3.4%) living with HIV and accounting for more than two-thirds of the people living with HIV worldwide [3].Ghana has HIV prevalence of 1.7%, affecting 334,713 people and accounting for over fourteen thousand annual deaths [4].HIV testing has been a gateway to HIV care, however coverage with HIV testing services is not adequate [5].Despite the availability and accessibility of HIV testing, people continue to test late in the course of HIV infection [6].
In 2015, WHO recommended that all people living with HIV start ART (Antiretroviral therapy) irrespective of clinical or immune status [7].However, people living with HIV continue to present to care late and with advanced disease [7].Late presentation with Advanced HIV disease (LP-AHD) is defined as an individual presenting with a CD4 + count lower than 200 cells/µL or at WHO clinical stage-3 or stage-4 [8].LP-AHD can lead to serious outcomes, such as increased mortality [9,10], development of opportunistic infections [11], increased risk of drug resistance to antiretroviral therapies (ART) [12], high healthcare costs [12], and increased transmission due to ignorance of infection status [13].
Multiple sociodemographic, psychosocial, and structural risk factors at the patient and provider levels have been found to be associated with LP-AHD.Fear of HIVrelated stigma [14] and discrimination [15], poor social support [14], and low risk perception [13] are among the common patient-related factors that discourage people from seeking timely testing.Providers have described inadequate time and resources [16], the burdensome counseling and consent process [17], and the provider's perceived low risk of transmission [18], as barriers to providing HIV testing.
Despite the challenges LP-AHD poses to HIV care, there is a paucity of data on the prevalence and factors associated with LP-AHD.There is no study that has been done to assess the scope of LP-AHD in the Ghanaian context.For the first time, we determined the prevalence and predictors of LP-AHD among newly diagnosed HIV/ AIDS subjects in Ghana.

Study design and study site
A retrospective cross-sectional study was conducted at HIV clinics of Komfo Anokye Teaching Hospital (KATH) and Aniniwah Medical Centre (AMC) in the Ashanti Region of Ghana.KATH and AMC are located in Kumasi, the capital of the Ashanti region of Ghana and have a well-resourced HIV care clinics making it suitable for the successful implementation of the study.

Study population
This study recruited newly diagnosed individuals confirmed of HIV/AIDS by clinicians using the standard diagnoses protocols.

Inclusion criteria
Participants eligible for inclusion in this study were adults aged 18 years and older with a newly confirmed diagnosis of HIV infection based on standardized diagnostic criteria.In addition, these participants were willing to share relevant information for the research study.Also, these individuals had complete and valid data for the study variables, which encompass sociodemographic, clinical, and psychosocial factors considered in the study.

Exclusion criteria
Individuals below the age of 18 are excluded, as well as those with incomplete or missing data related to crucial study variables encompassing sociodemographic, clinical, and psychosocial factors.Additionally, individuals who decline informed consent or express unwillingness to share pertinent information for research purposes were not considered.

Sample size calculation and sampling technique
The sample size was calculated using the Cochrane formula: n = z 2 pq e 2 ; where: n is the minimum sample size, p is the prevalence of HIV in Ghana = 1.7% [6], q = 1-p, (98.3), z = z value at 95% confidence (1.96), and e is the margin of error (0.05).Hence a minimum of twenty-six (26) participants were required for the study.To increase statistical power, 315 people diagnosed of HIV/AIDS were included for the study.The participants included 252 from the Komfo Anokye Teaching Hospital (KATH) and 63 from the Aniniwah Medical Centre, both within the Ashanti Region of Ghana.

Conclusion
The prevalence of LP-AHD among PDWH in Ghana is high.Efforts to improve early initiation of HIV/AIDS care should focus on factors such as the high perceived costs of HIV care, diagnosis based on clinical suspicion, and missed opportunities for early diagnosis by physicians.Keywords Latent HIV, HIV/AIDS, Advance diseases, Self-testing

Ethical consideration
Ethical approval was sought from the Committee on Human Research, Publication, and Ethics at the School of Medicine and Dentistry of the Kwame Nkrumah University of Science and Technology (KNUST) (CHRPE/ AP/385/23).Written informed consent was sought from each participant before the data collection of the study.Participation was voluntary and participants were allowed to opt out any time on their personal reasons.

Data collection
Data for this study were collected using a structured questionnaire comprising four main sections: Sociodemographic, Lifestyle, Clinical Factors, and Psychosocial Factors.The questionnaire was designed to gather comprehensive information pertaining to the study's objectives.Data were collected through face-to-face interviews conducted by trained research assistants.Participants' responses were recorded as per their answers to the structured questions in the questionnaire.Participants were stratified into late presenters and non-late presenter based on the WHO HIV staging system.

Definition of dependent variables
Participant with LP-AHD -People confirmed with HIV/AIDS at WHO stage 3 or 4 at the time of diagnosis.
Participant without LP-AHD-People confirmed with HIV/AIDS at WHO stage 1 or 2 at the time of diagnosis.

Data management and statistical analyses
Data analysis was performed using Statistical Package for the Social Sciences 26.0 software (SPSS, Inc.; Chicago, IL, USA) and GraphPad Prism version 8.0.Categorical variables were presented as frequencies and percentages.A simple bar chart was used to illustrate the prevalence of LP-AHD among study participants.Pearson chi-square test or Fischer exact test was conducted to ascertain the relationship between LP-AHD and the study variables.Variables were also tested in the Univariate logistic regression prediction model and significant variables from the univariate and potential cofounders were tested in the multivariate logistic regression prediction model to assess the independent predictors of LP-AHD.p-value of less than 0.05 and 95% confidence interval were considered statistically significant.

Sociodemographic characteristics of study participants
Of the 315 participants enrolled in the study, one-third (33.3%) were within the ages of 46-55years, one-quarter (25.4%) were within the age range of 36-45 years with the least percentage of age range being 18-35 years (13.0%).Most of the participant in the study were females (85.7%) with males accounting for 14.3% of the population.Considering marital status, 37.5% were married with 17.5% being single.Also, a higher percentage of the participants had 2-3 children (44.3%) and Junior High School education (39.0%).Majority of the enrolled participants lived in urban areas (85.7%), were Christians (87.6%) and Akans (81.6%).With respect to employment status, 7.3% were employees whilst 63.5% were selfemployed.Moreover, 47.5% of the participants earned < C 500 and 45.0% took 31-60 min to reach the clinic on visit days (Table 1).

Prevalence of LP-AHD
The study found that the prevalence of LP-AHD among the study participants was 28.6% which accounted for 90 of the participants.225 of the study participants did not present late (71.4%) (Fig. 1).

Lifestyle factors associated with late presentation with advanced HIV disease (LP-AHD)
Alcohol intake (p value = 0.714), alcohol Intake Frequency (p value = 0.152), knowledge of partner's HIV status (p = 0.164), history of smoking (p = 0.798), action taken by participants when feeling sick (p = 0.167), history of Needle Sharing for Drug Use (p = 0.352), history of blood donation (p = 0.129) and number of sexual partners (p = 0.479) all showed no statistical significance with LP-AHD (Table 3).

Psychosocial factors associated with late presentation with advanced HIV disease
Various reasons for delayed testing (p < 0.001) showed a significant association with LP-AHD.Additionally, participants with a fear of perceived side effects of ART (p = 0.002), participants who expressed concerns about confidentiality (p = 0.002), had difficulty accessing healthcare (p = 0.029), fear of stigma associated with HIV/AIDS (p < 0.001), participants who disclosed their HIV Status to partner/close relative (p = 0.011), participants who sought medical care that eventually led to their HIV diagnosis (p < 0.0001) were all significantly associated with LP-AHD (Table 5).

Discussion
Although there has been remarkable progress in HIV prevention and treatment, LP-AHD is still a major public health problem globally.A significant proportion of people living with HIV do not know they are infected and therefore seek medical care late.This study revealed that approximately 29% (28.6%) of the study participants presented late for care at WHO HIV stages 3 and 4.Moreover, the age group of 36-45 years, perceived high cost of HIV care, diagnosis based on clinical suspicion, and missed opportunities for early diagnosis by were associated with LP-AHD.
The prevalence of LP-AHD in this study, 28.6% was lower compared to previous studies conducted by Gesesew et al., in Ethiopia (65.5%) [5] and Jeong et al., in Asia (72-83.3%)[19].The disparity could be due to the utilization of just WHO staging for defining LP-AHD in this study whilst the other studies employed both CD4 count and WHO staging in their classification.Although the prevalence of LP-AHD in this study is lower compared to the overall prevalence of late presentation in Africa estimated to be between 35% and 65%, it still undermines the 95-95-95 targets set by the UNAIDS to end HIV/AIDS by 2030.The current study did not find any significant association between gender and LP-AHD.However, there was a larger proportion of males (37.8%) who presented with LP-AHD compared to females (27.0%); even though majority of study population included in the study were females (85.7%).This could be due to the fact that men may not be seeking care consistently and are accessing treatment at a later stage of their disease than women, which has also been reported in previous study [20].Previous studies have cited being male as a strong correlate for LP-AHD [14,21,22].However this trend conflicts with other studies which showed that being female is a risk factor for LP-AHD [5,23].This study findings call for enhance public health education especially among males on the clinical important of continuous health seeking behaviors.
Moreover, participants in lower age group between 36 and 45 years bracket were less likely to present late than their older counterparts and this is supported by other studies which showed that old age was a predictor of LP-AHD [13,24].This may be due to the misconception that older people are perceived to be at lower risk of HIV infection, which may lead to lower awareness and a lower sense of urgency for regular HIV testing in older populations.
Also, seeking herbal medication was a significant cause of two-fold increased likelihood of LP-AHD among study participants.Reliance on unorthodox medication has also been reported by Agaba et al. as a cause of LP-AHD [25].Over dependence on herbal medicine and faith healers coupled with unaware of HIV status could account for HIV progression and eventual presentation with advanced HIV disease.
Investigating the type of HIV infection did not reveal a conclusive statistically significant relationship with LP-AHD.However, a previous study conducted in Guinea-Bissau showed that HIV-2 and HIV-1/2 dually infected patients had lower risk of late presentation compared with HIV-1 infected patients [26].Differences in study populations and methods may account for this disparity.While history of other STDs and opportunistic infection were not statistically significant findings in this current study, results from a prospective study of 115 PDWH in Turkey [27] showed that opportunistic infections were likely implications of late diagnosis as the immune system is weakened and more susceptible to other infections.People with such infections fall into the WHO HIV staging criteria for late presentation and are therefore diagnosed as such.
Although there is routine HIV testing in the Ghanaian healthcare setting, individuals may not be screened for  HIV unless they specifically request a test or present with symptoms that prompt clinicians to consider HIV testing.This finding ties in with the missed chances of diagnosis observed in this study as patients were not tested initially when they reported to the hospital with symptoms.Furthermore, those who perceived HIV testing as expensive showed a nineteen times increased risk of LP-AHD.This calls for awareness creation as HIV testing is covered by the Ghana national health insurance [28].Furthermore, concerns about confidentiality were associated with more than a two-fold increased risk of LP-AHD although it didn't reach significance after adjusting for cofounders.Other studies have also identified confidentiality issues as a barrier to HIV self-testing [13,29] showing the importance of privacy protection and adherence to ethical standards in the healthcare setting.
The findings from this study could inform policymakers, public health officials, and HIV physicians in their fight against HIV.By recognizing the challenges to a timely HIV diagnosis, physicians can prioritize HIV testing, especially for people who show evidence of symptoms or risk factors.Additionally, awareness of the association between the perceived cost of HIV treatment and delayed presentation may prompt physicians and public health officials to address financial misinformation regarding HIV care since WHO has been HIV care entirely free as well increase access to testing and treatment options.
This study presents a novel data on the prevalence and predictors of LP-AHD in Ghana.However, using WHO HIV grading alone to distinguish between late and nonlate presenters limited the scope of this study.Additionally, our study used participant self-reported responses and available hospital records, which presented issues with subject bias and data completeness.The retrospective nature of this study also shows the need for a prospective investigation that would allow for real-time monitoring of late presenters and, also to reduce the possibility of bias.Future studies are needed to investigate the immunological correlates of late presenters.

Conclusion
The prevalence of LP-AHD among people diagnosed with HIV is high.Missed opportunities for early diagnosis by clinicians, diagnosis based on clinical suspicion,  old age and perceived cost of HIV care were associated with LP-AHD.These factors call for targeted campaigns among the general populace and clinicians alike to promote routine HIV testing and enhanced education on the awareness of early signs of HIV.

Fig. 1
Fig. 1 Prevalence of late presentation with advanced HIV disease (LP-AHD)

Table 1
Sociodemographic characteristics of study participants Data presented as frequency and percentage.

Table 2
Sociodemographic factors associated with late presentation with advanced HIV disease (LP-AHD) Note: Data presented as frequency and percentage, p-values computed byChi-square, p < 0.05 and bolded means statistically significant, LP-AHD: Late presentation with advanced HIV disease

Table 3
Lifestyle factors associated with LP-AHD Note:

Table 4
Clinical factors associated with LP-AHD Note:

Table 5
Psychosocial factors associated with LP-AHD Note:

Table 6
Predictors of LP-AHD Note: